JPH01109202A - Visual sensor with parallax correction - Google Patents
Visual sensor with parallax correctionInfo
- Publication number
- JPH01109202A JPH01109202A JP26523787A JP26523787A JPH01109202A JP H01109202 A JPH01109202 A JP H01109202A JP 26523787 A JP26523787 A JP 26523787A JP 26523787 A JP26523787 A JP 26523787A JP H01109202 A JPH01109202 A JP H01109202A
- Authority
- JP
- Japan
- Prior art keywords
- camera
- reference plane
- height
- optical axis
- reference surface
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000000007 visual effect Effects 0.000 title claims abstract description 12
- 230000003287 optical effect Effects 0.000 claims abstract description 17
- 238000010586 diagram Methods 0.000 description 7
- 241000282320 Panthera leo Species 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Abstract
Description
【発明の詳細な説明】
産業上の利用分野
本発明は、ロボット等の視覚として、作業対象を認識す
るために利用される視覚センサに関する。DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a visual sensor that is used as a visual aid for a robot or the like to recognize a work object.
従来の技術
ロボット等の視覚として利用される視覚センサにおいて
は、カメラで踊彰した画像の中の対象は奥行きのない二
次元平面の位置で認識される。そして、従来は、カメラ
と認識対象物間の距離は常に一定にして対象物の位置を
検出するようにしている。In conventional visual sensors used for vision in robots and the like, objects in images captured by a camera are recognized at positions on a two-dimensional plane with no depth. Conventionally, the distance between the camera and the object to be recognized is always kept constant to detect the position of the object.
発明が解決しようとする問題点
カメラで認識しようとする対象物の高さが変わると、カ
メラで撮影した画面上の対象物の位置は変動する。第６
図は対象物の高さが変ったとき、カメラが認識する対象
物の位置を説明する図で、Ｘ軸上を基準面とし、該基準
面よりＨｌだけ高い所、即ら、カメラ１側にある対象物
Ｐ１は、基準面上では×１の位置に存在し、カメラ１の
撮影画面にも×１の位置に画像があるべきにもかかわら
ず、カメラ１で撮影した画面にはＸ１’の位置に画像が
ｌｌａ影されることとなる。また、カメラ１に対し、基
準面より遠い位置Ｈ２、即ち、低い所にある対象物Ｐ２
はカメラ画面上に×２の位置にあるべぎにもかかわらず
、Ｘ２’ の位置に存在することとなる。このように、
カメラ１と対象物間の距離、即ら、視差によって対象物
を認識する位置が変動する。Problems to be Solved by the Invention When the height of the object to be recognized by the camera changes, the position of the object on the screen photographed by the camera changes. 6th
The figure explains the position of the object recognized by the camera when the height of the object changes.The reference plane is on the X-axis, and the position is Hl higher than the reference plane, that is, on the camera 1 side. A certain object P1 exists at the position x1 on the reference plane, and even though the image taken by camera 1 should also be located at the position x1, the screen taken by camera 1 shows an image of x1'. The image will be shadowed at the position. Also, with respect to the camera 1, an object P2 located at a position H2 farther from the reference plane, that is, a lower place
exists at position X2' even though it is located at position x2 on the camera screen. in this way,
The position at which the object is recognized varies depending on the distance between the camera 1 and the object, that is, the parallax.
そのため、従来は、カメラと対象物間の距離を一定に保
持して対象物の位置を検出し、対象物の高さ（カメラと
対象物間の距離）が変われば、カメラの高さ位置を変え
、所定距離にする必要があった。Therefore, in the past, the distance between the camera and the object was kept constant to detect the position of the object, and if the height of the object (distance between the camera and the object) changed, the height position of the camera was detected. It was necessary to change the distance and set it to a specified distance.
そこで、本発明の目的は、認識対象物の高さが変り、カ
メラの視差が変っても、視差補正を行って対象物の位置
を正確に検出できる視覚センサを提供することにある。SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a visual sensor that can perform parallax correction and accurately detect the position of a recognition target even if the height of the recognition target changes and the camera parallax changes.
問題点を解決するための手段
第１図は、本発明が上記問題点を解決するために採用し
た手段のブロック図で、本発明は、カメラから基準面ま
での光軸距離及び認識対象物の基準面からの高さを入力
設定するための入力手段と、該入力手段から入力された
設定値を記憶する記憶手段と、カメラが撮影した認識対
象物の両僅位置及び上記記憶手段に記憶する上記光軸距
離、対象物の基準面からの高さより認識対象物の基準面
平面における位置を求める視差補正手段とを設けること
によって視差を補正して認識対象物の位置を検出し、上
記問題点を解決した。Means for Solving the Problems FIG. 1 is a block diagram of the means adopted by the present invention to solve the above problems. an input means for inputting and setting the height from the reference plane; a storage means for storing the setting value inputted from the input means; and a storage means for storing the position of the recognition target photographed by the camera and in the storage means. By providing a parallax correction means that calculates the position of the recognition target on the reference surface plane from the optical axis distance and the height of the target from the reference plane, the position of the recognition target can be detected by correcting the parallax, and the above problem can be solved. solved.
作 用
第２図、第３図は、本発明の原理作用を説明する説明図
で、第２図において、Ｘ軸上を基準面とし、基準面より
Ｈだけ高い所（カメラ側）に対象物Ｐが存在するとする
。カメラ１の基準面からの距離がし、カメラの光軸とＸ
軸（基準面）との交点を原点とし、対象物位ｆａＰから
光軸及びＸ軸に下ろした垂線の光軸、Ｘ軸との交点をｐ
’ 、ｘ。Operation Figures 2 and 3 are explanatory diagrams explaining the principle operation of the present invention. In Figure 2, the reference plane is on the Suppose that P exists. The distance from the reference plane of camera 1 is
The point of intersection with the axis (reference plane) is the origin, and the point of intersection with the optical axis and the X-axis of a perpendicular line drawn from the object position faP to the optical axis and the X-axis is p.
', x.
及びカメラから対象物位置Ｐを通る直線のＸ軸との交点
をＸ′とすると、カメラ１が対象物Ｐを撮影したときカ
メラ画面には、対象物ＰはＸ′の位置にあるものとして
検出される。そこで、真の対象物Ｐの基準面位ｔｉｆｆ
Ｘは三角形
ＬＰ’Ｐと三角形ＬＯＸ’が相似であるから、上記第（
１）式によって求められる。And if the intersection point of the straight line passing from the camera through the object position P with the be done. Therefore, the reference plane position tiff of the true object P
Since triangle LP'P and triangle LOX' are similar to X, the above (
1) It is determined by the formula.
また、第３図に示すように、対象物Ｐが基準面より低い
位置（カメラ１より遠い位置）にあると、カメラ１が検
出する対象物Ｐの位置はＸ′の位置であり、この場合の
対象物Ｐの真の位置Ｘは三角形ＬＸ’ Ｏと三角形ｐｘ
’ ｘが相似であることから
となる。そこで、基準面よりカメラ１に近い側に対象物
Ｐがあるときは、その位置を基準面との距１ＩＩ（高さ
）Ｈをプラスで、また、遠い側にあるときの基準面と対
象物８間の距離Ｈをマイナスで設定すると、対象物Ｐの
真の基準面上の位置Ｘは第（１）式で求まることとなる
。Furthermore, as shown in Fig. 3, when the object P is located at a lower position than the reference plane (farther than the camera 1), the position of the object P detected by the camera 1 is the position X', and in this case The true position X of the object P is the triangle LX' O and the triangle px
' This is because x is similar. Therefore, when the object P is on the side closer to camera 1 than the reference plane, its position is set by adding the distance 1II (height) H from the reference plane, and when it is on the far side, the distance between the reference plane and the object If the distance H between 8 and 8 is set as a negative value, the true position X of the object P on the reference plane can be found by equation (1).
また同様に、基準面のＹ軸座標位置も求められ、結局、
対象物Ｐの基準面上の位置は
として求められることとなる。Similarly, the Y-axis coordinate position of the reference plane is also found, and eventually,
The position of the object P on the reference plane is determined as follows.
そこで、上記入力手段よりカメラ１から基準面までの光
軸距離し、対象物Ｐの基準面からの距離（高さ）Ｈを入
力設定し、記憶手段に記憶させておき、カメラ１で対象
物Ｐを撮影し、対象物の見かけ上の位１（Ｘ’ 、Ｙ’
）を求めて上記第（１）式、第（２）式の演算を視差
補正手段で行い、真の対象物位置を求めるこ−とができ
る。Therefore, the optical axis distance from the camera 1 to the reference plane is input and set using the input means, and the distance (height) H of the object P from the reference plane is input and set in the storage means. P, and the apparent order of the object (X', Y'
) and calculate the above equations (1) and (2) using the parallax correction means, thereby making it possible to find the true position of the object.
実施例
第４図は、本発明の一実施例の視覚センサの要部ブロッ
ク図である。Embodiment FIG. 4 is a block diagram of essential parts of a visual sensor according to an embodiment of the present invention.
図中１は基準面１０の上方に配設されたカメラ、Ｐは認
識対象物を表わす、２は中央処理装置（以下、ＣＰＵと
いう）で、３はカメラインターフェイス４を介して入力
されたカメラ１が撮影した基準面１０の画像を記憶する
フレームメモリ、５はデータメモリで、後述するように
、手動データ入力装置７から入力される基準ｖａ１０と
カメラ１間の光軸上の距離Ｌ１及びＷ１識対象物の高さ
Ｈ等をパラメータとして記憶するものである。また、６
はプログラムメモリで、ＣＰＵ２が実行するｉ制御プロ
グラムを記憶するものである。なお、８はバスである。In the figure, 1 is a camera disposed above the reference plane 10, P represents a recognition target, 2 is a central processing unit (hereinafter referred to as CPU), and 3 is a camera 1 input via the camera interface 4. 5 is a data memory that stores the image of the reference surface 10 photographed by the camera 1, and 5 is a data memory that stores the distances L1 and W1 on the optical axis between the reference va10 input from the manual data input device 7 and the camera 1. The height H of the object and the like are stored as parameters. Also, 6
is a program memory that stores the i control program executed by the CPU 2. Note that 8 is a bus.
以上のような構成において、カメラ１から基準面１０ま
での光軸距離りを測定し、手動データ入力装置より該光
軸距離りを入力し、データメモリ５内に記憶させてお（
。さらに、ｉ［しようとする各対象物の基準面１０から
の距離（高さ）Ｈも手動データ入力装置７より入力し、
データメモリ５内にパラメータとして記憶させるが、こ
の場合、対象物Ｐの上面が基準ＷＪ１０より突出し、高
い場合は、基準面と対象物Ｐの上面間の距離（高さ）Ｈ
をプラスの値で設定する（通常この場合が多い）。また
、対象物Ｐの上面位置が基準面１０よりもカメラ１から
遠い場合には、基準面と対象物Ｐの上面位置間の距離Ｈ
を高さとしてマイナス符号で設定する。In the above configuration, the optical axis distance from the camera 1 to the reference plane 10 is measured, the optical axis distance is inputted from the manual data input device, and is stored in the data memory 5 (
. Furthermore, the distance (height) H of each target object from the reference plane 10 is also input from the manual data input device 7,
It is stored as a parameter in the data memory 5. In this case, if the top surface of the object P protrudes and is higher than the reference WJ10, the distance (height) between the reference surface and the top surface of the object P
Set to a positive value (usually this is the case). In addition, if the top surface position of the object P is farther from the camera 1 than the reference plane 10, the distance H between the reference surface and the top surface position of the object P
Set as the height with a minus sign.
そこで、検出しようとする対象物Ｐの高さＨを手動デー
タ入力装置７で選択設定した後、該視覚センサを稼動さ
せてカメラ１より画像をカメラインターフェイス４を介
してフレームメモリ３内に取込む。フレームメモリ３中
のカメラ画面の対象物Ｐの画像が第５図に示すように画
素位置（Ｘ′。Therefore, after selecting and setting the height H of the object P to be detected using the manual data input device 7, the visual sensor is activated to capture an image from the camera 1 into the frame memory 3 via the camera interface 4. . The image of the object P on the camera screen in the frame memory 3 is located at the pixel position (X') as shown in FIG.
Ｙ’ ）で検出されたとすると、該検出位置（Ｘ’。If it is detected at the detection position (X').
Ｙ’ ）より真の対象物Ｐの位置（Ｘ、Ｙ）を求めるこ
ととなるが、フレームメモリ３中のカメラ画面の光軸位
置、即ち、カメラ画面の中心位置を（ＸＯ、ＹＯ’）と
すれば、ＣＰＵ２はこのカメラ画面中心位ＩＩ （ＸＯ
、ＹＯ）とデータメモリ５に設定記憶され、選択された
＾さＨ９光軸距１１１Ｌ及び検出された画像位置（Ｘ’
、Ｙ’ ）より、次の第（４）式、第（５）式の演算
を行って真の座標位置（Ｘ、Ｙ）を求める。The true position (X, Y) of the object P is determined from (X, Y'), but the optical axis position of the camera screen in the frame memory 3, that is, the center position of the camera screen is determined as (XO, YO'). Then, CPU2 will move to the center of this camera screen II (XO
, YO) are set and stored in the data memory 5, and the selected ^^ optical axis distance 111L and the detected image position (X'
, Y'), calculate the following equations (4) and (5) to find the true coordinate position (X, Y).
なお、第（４）式、第（５）式と第（１）式、第（３）
式と相違する点は、第（１）式、第（３）式においては
、光軸座標位置が第２図、第３図に示すように原点とし
ていたが、この光軸座標位置が第５図に示すように、（
ＸＯ，ＹＯ）にシフトしている関係から座標変換を行う
ためである。In addition, equations (4), (5), equations (1), and (3)
The difference from the formula is that in formulas (1) and (3), the optical axis coordinate position is the origin as shown in Figures 2 and 3, but this optical axis coordinate position is the origin as shown in Figures 2 and 3. As shown in the figure, (
This is to perform coordinate transformation based on the relationship shifted to (XO, YO).
こうして、ＣＰＵ２は第（４）式、第（５）式の演算を
行って対象物Ｐの真の位１（Ｘ、Ｙ）を求めることがで
き、求められた位ＩＩ（Ｘ、Ｙ）をロボット等に出力す
ればよいこととなる。In this way, the CPU 2 can calculate the true digit 1 (X, Y) of the object P by calculating the equations (4) and (5), and the obtained digit II (X, Y). All you have to do is output it to a robot or the like.
発明の効果
本発明は、視覚センサが視差補正を行うため、高さの異
なる対象物に対しても、簡単に、かつ、正確にその位置
を検出することができる。Effects of the Invention In the present invention, since the visual sensor performs parallax correction, the position of objects of different heights can be detected easily and accurately.
第１図は、本発明が従来技術の問題点を解決するために
採用した手段のブロック図、第２図、第３図は、本発明
の原理９作用を説明する説明図、第４図は、本発明の一
実施例の要部ブロック図、第５図は、同実施例における
フレームメモリ上のカメラ画面の説明図、第６図は、従
来の視覚センサが検出する高さが異なる対象物の位置の
説明図である。
１・・・カメラ、２・・・中央処理装置（ＣＰＵ）、３
・・・フレームメモリ、４・・・カメラインターフェイ
ス、５軸・データメモリ、６軸・プログラムメモリ、７
・・・手動データ入力装置、８・・・パス、９・・・対
象物、１０・・・基準面、し・・・カメラから基準面ま
での光軸距離、トド・・対家物高さ。
第 １ 口
第 ３図 第２囚FIG. 1 is a block diagram of the means adopted by the present invention to solve the problems of the prior art, FIGS. 2 and 3 are explanatory diagrams explaining the principle 9 operation of the present invention, and FIG. , a main part block diagram of an embodiment of the present invention, FIG. 5 is an explanatory diagram of a camera screen on a frame memory in the same embodiment, and FIG. 6 is an illustration of objects of different heights detected by a conventional visual sensor. It is an explanatory view of the position of. 1... Camera, 2... Central processing unit (CPU), 3
...Frame memory, 4...Camera interface, 5 axes/data memory, 6 axes/program memory, 7
...Manual data input device, 8...Path, 9...Target, 10...Reference plane, Shi...Optical axis distance from camera to reference plane, Sea lion...Height relative to house . 1st mouth 3rd prisoner 2nd prisoner
Claims (2)
の基準面からの高さを入力設定するための入力手段と、
該入力手段から入力された設定値を記憶する記憶手段と
、カメラが撮影した認識対象物の画像位置及び上記記憶
手段に記憶する上記光軸距離、対象物の基準面からの高
さより認識対象物の基準面平面における位置を求める視
差補正手段とを有し、視差を補正して認識対象物の位置
を検出するようにした視差補正付視覚センサ。(1) input means for inputting and setting the optical axis distance from the camera to the reference plane and the height of the recognition target from the reference plane;
a storage means for storing the setting values inputted from the input means; and a storage means for storing the setting values input from the input means, and a recognition target based on the image position of the recognition target taken by the camera, the optical axis distance stored in the storage means, and the height of the target from the reference plane. 1. A visual sensor with parallax correction which detects the position of a recognition target by correcting the parallax.
の基準面からの高さＨ、カメラ撮影画面の中心座標位置
を（Ｘ０、Ｙ０）、対象物のカメラ撮影画面の画像位置
を（Ｘ′、Ｙ′）とすると、上記視差補正手段は次式の
演算を行って、 Ｘ＝（Ｌ−Ｈ）／Ｌ・（Ｘ′−Ｘ０）＋Ｘ０Ｙ＝（Ｌ−
Ｈ）／Ｌ・（Ｙ′−Ｙ０））＋Ｙ０基準面平面における
対象物の座標位置（Ｘ、Ｙ）を求める特許請求の範囲第
１項記載の視差補正付視覚センサ。(2) The optical axis distance L from the camera to the reference plane, the height H of the object to be recognized from the reference plane, the center coordinate position of the camera shooting screen (X0, Y0), and the image position of the object on the camera shooting screen. (X', Y'), the parallax correction means calculates the following formula to obtain X=(L-H)/L・(X'-X0)+X0Y=(L-
A visual sensor with parallax correction according to claim 1, which determines the coordinate position (X, Y) of an object on a reference surface plane.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP26523787A JPH01109202A (en) | 1987-10-22 | 1987-10-22 | Visual sensor with parallax correction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP26523787A JPH01109202A (en) | 1987-10-22 | 1987-10-22 | Visual sensor with parallax correction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH01109202A true JPH01109202A (en) | 1989-04-26 |
Family
ID=17414427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP26523787A Pending JPH01109202A (en) | 1987-10-22 | 1987-10-22 | Visual sensor with parallax correction |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH01109202A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0392712A (en) * | 1989-09-05 | 1991-04-17 | Fanuc Ltd | Three-dimensional position recognition by use of image processing device and distance measuring sensor |
| JP2011007872A (en) * | 2009-06-23 | 2011-01-13 | Olympus Corp | Observation device and magnification correction method |
| JP2019098409A (en) * | 2017-11-28 | 2019-06-24 | 東芝機械株式会社 | Robot system and calibration method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5896204A (en) * | 1981-12-04 | 1983-06-08 | Kawasaki Steel Corp | Measuring method for size of material to be rolled |
| JPS6110708A (en) * | 1984-06-25 | 1986-01-18 | Kawasaki Steel Corp | Size measuring device using measuring camera |
-
1987
- 1987-10-22 JP JP26523787A patent/JPH01109202A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5896204A (en) * | 1981-12-04 | 1983-06-08 | Kawasaki Steel Corp | Measuring method for size of material to be rolled |
| JPS6110708A (en) * | 1984-06-25 | 1986-01-18 | Kawasaki Steel Corp | Size measuring device using measuring camera |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0392712A (en) * | 1989-09-05 | 1991-04-17 | Fanuc Ltd | Three-dimensional position recognition by use of image processing device and distance measuring sensor |
| JP2011007872A (en) * | 2009-06-23 | 2011-01-13 | Olympus Corp | Observation device and magnification correction method |
| JP2019098409A (en) * | 2017-11-28 | 2019-06-24 | 東芝機械株式会社 | Robot system and calibration method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11625842B2 (en) | Image processing apparatus and image processing method | |
| KR900002509B1 (en) | Apparatus for recognizing three demensional object | |
| KR101379787B1 (en) | An apparatus and a method for calibration of camera and laser range finder using a structure with a triangular hole | |
| JP3842988B2 (en) | Image processing apparatus for measuring three-dimensional information of an object by binocular stereoscopic vision, and a method for recording the same, or a recording medium recording the measurement program | |
| Li et al. | A calibration algorithm of the structured light vision for the arc welding robot | |
| JP2730457B2 (en) | Three-dimensional position and posture recognition method based on vision and three-dimensional position and posture recognition device based on vision | |
| JP2018173882A (en) | Information processing device, method, and program | |
| JPH01109202A (en) | Visual sensor with parallax correction | |
| KR100773271B1 (en) | Method for localization of mobile robot with a single camera | |
| KR101438514B1 (en) | Robot localization detecting system using a multi-view image and method thereof | |
| JPH02183102A (en) | Automatic calibration adjusting system for visual sensor | |
| CN113733078B (en) | Method for interpreting fine control quantity of mechanical arm and computer-readable storage medium | |
| Brandt et al. | Controlled active exploration of uncalibrated environments | |
| WO2021145304A1 (en) | Image processing system | |
| JPH09329440A (en) | Coordinating method for measuring points on plural images | |
| JPH04269194A (en) | Plane measuring method | |
| CN115205129A (en) | Depth camera based on structured light and method of use | |
| CN113324538A (en) | Cooperative target remote high-precision six-degree-of-freedom pose measurement method | |
| JPH06214622A (en) | Work position sensor | |
| JPH06258028A (en) | Method and system for visually recognizing three dimensional position and attitude | |
| JPH05337785A (en) | Grinding path correcting device of grinder robot | |
| JPS60183509A (en) | Visual sense device | |
| KR100784734B1 (en) | Error compensation method for the elliptical trajectory of industrial robot | |
| TW448289B (en) | Image aided 3D curved surface re-establishment and curved surface area measuring system | |
| JPH0731536B2 (en) | Teaching data correction robot |